Slashdot Mirror
NVIDIA Launches New SLI Physics Technology
Thomas Hines writes "NVIDIA just launched a new SLI Physics technology. It offloads the physics processing from the CPU to the graphics card. According to the benchmark, it improves the frame rate by more than 10x. Certainly worth investing in SLI if it works."
www.gpgpu.org
This neither requires SLI nor is it limited to NVIDIA chips. NVIDIA is just launching it publicly. ATI will be showing it off behind closed doors this week.
Don't forget that http://www.ageia.com/ is already doing this, and set to ship their cards sometime this year hopefully. Of course the significant difference between the two is that you would only have to buy one card for the SLI solution.
It doesn't, according to the article.
"Quoting yourself is stupid." -Me
The one linked is a little bland for my taste
h ysics/
r g.mozilla%3Aen-US%3Aofficial_s&hl=en&percentage_se rved=100&tab=wn&ie=ISO-8859-1&q=NVIDIA+SLI+Physics &btnG=Search+News
this one is better:
http://www.tgdaily.com/2006/03/20/nvidia_sli_forp
Or choose you own adventure via Google news:
http://news.google.ca/news?client=firefox-a&rls=o
Well, I for one, want to have a smarter AI in all games. Unloading the "mundane" physics engine to the graphic card will hopefully spare more CPU cycles for the AI. After all, it's not graphics that matter in games. It's the gameplay.
--
Error 500: Internal sig error
given than GPU->CPU readbacks are a notorious perfomance killer.
That has not been true for a long long time. Since PCIe became a standard, bidirectional communication between CPUs and GPUs has been as easy as unidirectional communication.
LL
Soviet Russia was very technically progressive. While being bad for the consumer, as it were, communism or socialism isn't necessarily bad for innovation.
un burrito me trampeó.
Havok is the primarily used API in the gaming industry. It is the one being targeted by this implementation.
That said, it would presumably be possible to implement other APIs (if there is sufficient demand), given that the GPU hardware is now general enough to handle that level of computation.
LL
The article compared fancy physics effects on the CPU at ~6fps and fancy physics effects on the GPU at ~60fps. This is completely understandable. It does nothing for current games and you most definitely will not see framerates of 600.
I don't mean to flame, but how did you put Voodoo cards in a Quadra? They never made NuBus Voodoo cards, only PCI. Perhaps you mean a PowerMac of some sort?
Clearly, you misunderstand how cinematic 3D is rendered
Desktop GPUs will always be inferior to cinematic 3D, simply because cinematic 3D is rendered at a rate of several frames per day by a multi-million dollar farm of computers, while desktop GPUs must deliver dozens of frames per second all by itself.
A peek at what it took to render The Incredibles:
And again -- even this much hardware generated images measured in frames per day -- nowhere near the ~24 frames per second you'd want for real-time imaging. In fact, according to pixar.com it takes 6 - 90 hours to render one frame.
Uh...that's what one would think. But in reality, the readback performance is only between 450MB/s (OGL) and 900MB/s (DX), nowhere near the limit of the PCIE bus (you can check the GPGPU forums for these numbers). This is actually only about 2X faster than in the AGP 8X days.
h ighlight=read+bandwidth)
IIRC, as it stands, uploading to the graphics card is about 4X as fast as downloading from the graphics card. So yes, GPU->CPU is still a performance killer, contrary to what you think or believe. (for your reference, here's a quick link to one of the posts, which is agreed upon from some of the site admins: http://www.gpgpu.org/forums/viewtopic.php?t=2092&
Lots of other things use PCI-Express including:
Single and Dual Port 4X SDR and DDR InfiniBand over PCI-Express x8
Dual port 2Gb and 4Gb FibreChannel over PCI-Express x4
Ethernet (multiport 1 gigabit and 10 gigabit), over PCI-Express x4
Multi port FireWire 800 over PCI-Express x1
DualChannel UltraSCSI320 over PCI-Express x1
There are more probably... PCI-Express grew out of InfiniBand. They cut out the networking to make it cheaper for just inside a single system. Ironically, they put a lot of the networking back in for Advanced Switching Interconnect.
-- soldack
And it's the proper way of doing things.
Back in the day when Pc's were properly designed the modem, Hard drive controller, video , sound, and other cards all offloaded their work ON the card freeing up that work from the CPU.
Then manufacturers wanted to save money by making "soft" hardware. Winmodems, soundcards, cheap IDE controllers and other cheaper cards came on the market that used CPU cycles and even system RAM for thier functions forcing a fast machine to crawl doing the work that the prephrials should have been doing in the first place.
Most consumer crap is utter crap because of the driver doing most of the work in the host OS/PC. making your shiny new dual core 5ghz machine as slow as a older box with real hardware on it.
Read the article... they state that right now you can not share both graphics rendering and physics emulation on the same GPU, though they do plan to work on this in the future.
For now you need two GPUs, one for graphics, one for physics.
Fortunately for you, the errors at this scale will be less than 6E-8 of the most significant digit. An HL/HL2 map is usually scaled in units between -4000 and 4000, so your error might be about 0.00024. The player model is less precise than this. The hit box is even less precise. You will incur more error simply due to the fact that your mouse cursor only moves by a single pixel increment at a time which could be significant at a low resolution. In short, you missed because you can't aim, or because you lagged. If I were you, I'd yell LAGGGGGG. A lot. Over and over.
Actually, the US is Linus's home, as it is for most Linux-based industry as well (think Novell, RedHat, and that little cash register company with a blue logo). I would also hardly say that Nokia makes Finland comparable with the Korea, Japan, and the US when it comes to technology.
Don't worry about it though, I'm just spouting off American propaganda.
Who needs Ford, GM, IBM, Apple, HP, Microsoft, Intel, etc? Finland has Nokia!
Samsung, LG, and Hyundai? They are no Nokia!
Sony, Panasonic, Toyota, and Honda don't even offer a Linux-based mini-tablet!
Wow, you are right. Who would want capitalism? It's not like I just took a one-week vacation in the Bahamas, go out to eat several nice meals per week, have excellent medical care, a new car, and decent clothes all with lower-middle class earnings.
Seriously, you can say I am soft, greedy, etc., but there is no way in hell you will convince me or the 90% or so people in the US in similar or better conditions than me that we should want socialism.
Our approach produces better-looking movement than the low-end physics packages. We don't have the "boink problem", where everything bounces as if it were very light. Heavy objects look heavy. Our physics has "ease in" and "ease out" in collisions, as animators put it, derived directly from the real physics. When we first did this, back in the 200MHz era, it was slow for real time (a two-player fighter was barely possible) but now, game physics can get better.
Take a look at our videos. Few if any other physics systems can even do the spinning top correctly, let alone the hard cases shown.